Config
Log for #openttdcoop.devzone on 10th May 2011:
Times are UTC Toggle Colours
01:13:22  *** Lakie has quit IRC
07:26:50  <Brot6> OpenGFX+ Industries - Bug #2613: Missing "Conflicting industries" property (Terkhen) @ http://dev.openttdcoop.org/issues/2613#change-6679
07:30:11  *** andythenorth has joined #openttdcoop.devzone
07:30:27  <Terkhen> planetmaker^ for when you have time
07:34:31  <planetmaker> I don't expect to have much time today, I'm afraid
07:35:26  <Terkhen> no need to hurry :)
08:01:49  *** andythenorth has quit IRC
10:48:33  *** KenjiE20 has joined #openttdcoop.devzone
11:41:18  *** ODM has joined #openttdcoop.devzone
11:54:10  *** ODM has quit IRC
11:56:10  *** ODM has joined #openttdcoop.devzone
12:59:54  *** andythenorth has joined #openttdcoop.devzone
13:22:29  <Brot6> OpenGFX+ Industries - Feature #2615 (New): Different options for calculating station ratings (Terkhen) @ http://dev.openttdcoop.org/issues/2615
14:02:43  <Brot6> FISH - Revision 623:6791219ac9fa: Fix: lighting was wrong for small trader (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/6791219ac9fa
14:02:43  <Brot6> FISH - Revision 624:5ca6ee5229a5: Feature: improved appearance of medium utility fishboat (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/5ca6ee5229a5
14:03:57  <Brot6> FISH - Bug #2588 (Closed): Lighting wrong on Traders (andythenorth) @ http://dev.openttdcoop.org/issues/2588#change-6680
14:20:09  <planetmaker> Terkhen, wrt #2613: looks fine. I guess there's no better way or place to define the conflicting industry types unless one wants to duplicate the if...then...else chains which would be a PITA.
14:20:10  <Brot6> planetmaker: Terkhen: #2613 is http://dev.openttdcoop.org/issues/show/2613 "OpenGFX+ Industries - Bug #2613: Missing "Conflicting industries" property - #openttdcoop Development Zone"
14:20:45  <planetmaker> (yes, I procrastinated ;-) - it's too warm here :S )
14:34:21  *** Lakie has joined #openttdcoop.devzone
14:39:21  <Terkhen> yes, it is as simple as I could do it :)
14:39:29  <Terkhen> ok, I'll commit once I'm back at linux
15:30:14  <Yexo> http://devs.openttd.org/~yexo/grf2nml.diff current diff
15:30:22  <Yexo> http://devs.openttd.org/~yexo/firs.txt output for firs
15:30:50  <Yexo> known bug: overlapping action7's are not properly converted, but I don't plan to support those either
15:31:07  <Yexo> search for "// Sprite 33" to see it in action, a few lines are duplicated there and one error is unconditional
15:31:29  <andythenorth> can probably reimplement the currently small number of action 7's
15:31:40  <Yexo> no need for that
15:31:57  <Yexo> the converted code needs to be looked at manually anyway, so it's easy to fix it after converting
15:32:03  <andythenorth> hmm
15:32:04  <andythenorth> long
15:32:07  <andythenorth> :o
15:32:21  <Terkhen> callback_flags_low and callback_flags_high looks strange
15:32:32  <Yexo> 19805 lines, original nfo code was 62997 lines
15:32:41  <Terkhen> the code is nicer than I expected :)
15:32:43  <Yexo> Terkhen: correct, nml only has a single property
15:32:50  <Yexo> so conversion of those is not done
15:33:18  <andythenorth> Terkhen: FIRS isn't complicated :)
15:33:21  <andythenorth> just big
15:33:23  <Yexo> http://devs.openttd.org/~yexo/output.txt <- some output from the conversion script
15:33:44  <Terkhen> most of the work will be renaming and moving stuff to their right place, simpler than what I thought too
15:33:57  <andythenorth> we'd be able to output nfo from the nml and diff with firs nfo to try and find errors?
15:34:04  <Yexo> no
15:34:06  <andythenorth> or compile a grf from each, decompile and diff?
15:34:18  <Yexo> that would be closer, but also not directly possible
15:34:37  <Yexo> the nfo output from nml is not always as good as manually written nfo, and in some cases better
15:35:15  <Yexo> switch(FEAT_INDUSTRIES, SELF, action2_7015, STORE_PERM((65535 / 2), 0)) { <- nml will optimize the 65535 / 2 away for example
15:35:43  <Yexo> but it'll write inefficient (in number of actions) nfo code for: if (grf_future_status("Meo")) { error(FATAL, string(STR_ERROR_39)); }
15:35:43  <andythenorth> hmm
15:35:48  <andythenorth> how do we find errors?
15:35:57  <andythenorth> or rather, omissions
15:36:03  <Yexo> manually comparing the code and testing
15:36:30  <Yexo> most importantly all action1's and real action2's are not converted
15:36:43  <Yexo> since I'm taking the grf, not the nfo code as input
15:39:37  <planetmaker> hm... looks quite nice, Yexo :-)
15:39:43  <Terkhen> ah, that part sounds more complicated :P
15:40:02  <Yexo> to convert the action1's properly I'd have to read the nfo, not grf
15:41:59  <planetmaker> hm, preprocessed nfo - harder to read?
15:42:15  <Yexo> nfo is harder to read than grf
15:42:33  <planetmaker> what constraints would make it easier?
15:42:35  <Yexo> you almost can't get anything simpler than reading a grf file
15:42:41  <planetmaker> :-)
15:43:13  <Yexo> not sure, #954 has some code already but I've never tested that
15:43:13  <Brot6> Yexo: #954 is http://dev.openttdcoop.org/issues/show/954 "NewGRF Meta Language - Patch #954: NFO lexer and basic parser - #openttdcoop Development Zone"
15:43:43  <Yexo> it might even be possible to read most info from the grf and only the real sprites from the nfo
15:43:51  <Terkhen> if FIRS code uses some templates or regular code for them we might be able to convert them quickly by some script tailored to FIRS
15:44:01  <Yexo> that would make it a lot easier, since most escapes from the nfo code could just be ignored
15:44:02  <planetmaker> that's what I thought, Yexo :-)
15:44:57  <planetmaker> Terkhen, it uses many files... and industry / building sprites are not that easy to template
15:45:08  <planetmaker> they're somewhat unique by nature. Much more than vehicles
15:45:14  <Terkhen> ok, it was just some wishful thinking :P
15:45:40  <Yexo> hmm, grepping for "<number> sprites/graphics/" might already filter out only the real sprites
15:45:54  <planetmaker> I guess so
15:46:03  <Yexo> if that actually works the rest is quite easy
15:46:54  <Yexo> it looks like it does :)
15:46:59  <Terkhen> :)
15:47:23  <planetmaker> :-)
15:48:16  <andythenorth> Yexo: testing FIRS is very involved
15:48:30  <andythenorth> often it requires setting up cargo chains
15:48:41  <andythenorth> and savegames are of some use, but not 100%
15:48:54  <andythenorth> just fyi
15:48:57  <Yexo> andythenorth: it's not the little corner cases I'm worried about
15:49:03  <Yexo> for now it's just "getting the basics to work"
15:49:20  <Yexo> I'm pretty sure the varaction2 chains are already converted correctly
15:49:57  <andythenorth> Terkhen: what has been templated for nfo is in FIRS /templates
15:50:01  <andythenorth> there's a lot
15:50:18  <Yexo> that'll have to be redone for the nml version
15:50:22  <andythenorth> yarp
15:50:30  <planetmaker> there is indeed. It can and possibly should be re-done in nml
15:50:33  <andythenorth> this is an optimum point for conversion at least
15:50:39  <andythenorth> as I am not planning to ship much for some time
15:51:07  <planetmaker> but having those templates gives an idea already what is worthwhile to template
15:51:30  <andythenorth> possibly I should have branched FIRS trunk, and make the next version 0.6.5 (the nml version - 100% compatible with 0.6.4)
15:51:32  <Yexo> http://devs.openttd.org/~yexo/default.txt <- converted language file
15:51:53  <andythenorth> then (forward?) port new features to 0.7 and reimplement them in nml...
15:51:55  <planetmaker> andythenorth, you can still do that. No problem
15:52:07  <planetmaker> you can up to any previous version, branch and  backport some changes
15:52:10  <planetmaker> easy
15:52:14  <andythenorth> it could be a valid route
15:52:27  <andythenorth> as long as 'you / we /s' can be applied above
15:52:37  <planetmaker> :-)
15:53:30  <Terkhen> yes, before FIRS is "freezed" we can start checking the resulting nml code and decide on how to organize it
15:53:53  <planetmaker> I'm not sure there are waiting some bear traps with branches, but I *think* not as we did do that with FIRS once
15:53:54  <Terkhen> the best would be to mirror FIRS templates and organizations, but there might be some stuff that is better to change
15:54:12  <planetmaker> The work is to possibly back-port changes from trunk-FIRS to that branch. That might be work to do.
15:54:25  <Terkhen> the best example of stuff to change is translations :P
15:54:34  <planetmaker> :-)
15:54:43  <Terkhen> given the format used in FIRS, it should not be complicated to rewrite the script
15:54:54  <planetmaker> yep, indeed
15:54:55  <Yexo> rewrite which script?
15:55:00  <planetmaker> check_languages
15:55:10  <andythenorth> I'm not sure which is better route
15:55:15  <planetmaker> which checks the translations for being up to date
15:55:19  <Yexo> ah, that one
15:55:26  <andythenorth> (a) change one thing (nml) everything else is 100% compliant with FIRS 0.6.4
15:55:31  <Terkhen> Yexo: a helpful hack :)
15:55:41  <andythenorth> (b) change multiple things (nml + various changes that break savegames)
15:55:54  <andythenorth> I don't know which contains more risk
15:56:00  <Yexo> andythenorth: what state is the current version of the code?
15:56:07  <andythenorth> wrt 0.6.4?
15:56:14  <planetmaker> generally
15:56:22  <Terkhen> I'm not sure if the nfo->nml change will keep savegame compatibility
15:56:29  <Yexo> yes, is r1989 already savegame broken wrt 0.6.4?
15:56:31  <planetmaker> Terkhen, I'd not bet on that
15:56:38  <andythenorth> Yexo: in small ways
15:56:41  <Yexo> I don't see why it would not
15:56:46  <andythenorth> mostly wrt graphics
15:56:54  <planetmaker> different layouts?
15:57:08  <Yexo> since nml has the option of manually specifying IDs for action0, which the conversion uses
15:57:17  <planetmaker> well, ok :-)
15:57:36  <Terkhen> true, and for FIRS we should keep specifying them manually for compatibility anyways
15:57:37  <planetmaker> But it might be wise to not require compatibility between the nfo and nml version
15:57:42  <Yexo> item(FEAT_CARGOS, item_FEAT_CARGOS_1, 1) { <- last ", 1" is optional in pure nml code, but it keeps it compatible for the conversion
15:58:21  <Yexo> perhaps, but are there other reasons the compatibility would break by conversion?
15:58:29  <planetmaker> I think the main question is: should the current state be shipped as 0.7 or should the conversion be done before 0.7?
15:58:43  <planetmaker> Yexo, no intrinsic one. Parameters?
15:58:56  <andythenorth> if we change only nfo-nml and maintain compatibility, we stand a reasonable chance of getting bug reports which can be isolated to the nfo-nml switch
15:58:59  <planetmaker> But that's not a big issue either. Layout numbers?
15:59:03  <Yexo> parameter shouldn't be broken either
15:59:09  <Terkhen> andythenorth: that's a very good point
15:59:10  <Yexo> nor layout numbers
15:59:54  <planetmaker> then I see nothing which would break either.
16:00:01  <andythenorth> there are currently 12k installs of FIRS 0.6.4
16:00:09  <andythenorth> the adoption rate of new FIRS versions is quite slow
16:00:11  <Yexo> andythenorth: yes, but that would mean either releasing 0.6.5 (nml version) and waiting quite a while with 0.7 to see if any conversion bugs pop up
16:00:12  <Terkhen> neither do I, but with something as big as this there will be overlooks :)
16:00:20  <andythenorth> Yexo: I have no problem waiting for 0.7
16:00:24  <andythenorth> there are other things I can do
16:00:24  <Yexo> or releasing 0.7 now (at least before converting) and 0.7.1 afterwards
16:00:30  <Terkhen> or unforeseen problems
16:00:33  <planetmaker> Yexo, I'd not release any 0.6.x version with NML
16:00:43  <andythenorth> 0.7.0 could be the nml release
16:00:49  <andythenorth> but be 0.6.4 compatible
16:00:58  <andythenorth> 0.8 could be the compatibility-breaking release
16:01:09  <Yexo> andythenorth: as long as current head in hg is still compatible with 0.6.4, that'll work
16:01:16  <Terkhen> as long as the users can easily spot anything that changed the actual version do not matter IMO
16:01:25  <andythenorth> current head is *not* 0.6.4 compatible
16:01:26  <planetmaker> :-)
16:01:35  <andythenorth> I would convert 0.6.4 tag
16:01:38  <andythenorth> make that 0.7
16:01:41  <andythenorth> do some magic
16:01:46  <planetmaker> err... why?
16:01:49  <Yexo> andythenorth: converting 0.6.4 would mean doubling the work
16:01:53  <andythenorth> hmm
16:01:59  * andythenorth is not very good at vcs issues
16:02:04  <Terkhen> why not release as 0.7.0 and using 0.7.1 for the conversion?
16:02:12  <Yexo> as all changes since 0.6.4 would have to be ported to the nml version, or the conversion would have to be redone
16:02:20  <andythenorth> I would port the changes
16:02:25  <andythenorth> let me see
16:02:33  <Yexo> if you want to isolate the nml conversion, release 0.7.0 now with the changes you have
16:02:40  <Yexo> than 0.7.1 as nml conversion when it's ready (without other changes)
16:03:06  <V453000> sorry with interruption but which unix command do I see how many people are connected to the same @whatever as me?
16:03:33  <andythenorth> Yexo: current trunk can't be released.  That's the missing information here ;)
16:03:46  <Yexo> V453000: try "users"
16:03:50  <V453000> ooh :D
16:03:51  <V453000> thx
16:04:18  <planetmaker> andythenorth, why not?
16:04:31  <Yexo> andythenorth: that's certainly important missing information :)
16:05:43  <planetmaker> otoh: why would 0.7 need to be released in order to convert to nml?
16:06:11  <planetmaker> can't we just go from where we are and convert?
16:06:32  <planetmaker> or 'just' for the sake of easily comparing releases' behaviour?
16:06:56  <Terkhen> that would be quite useful yes
16:07:51  <Yexo> andythenorth: how much work would it be to finalize the 0.7 release?
16:07:54  <planetmaker> doubtless, yes
16:09:27  <andythenorth> Yexo: a lot
16:09:36  <andythenorth> my aim is to only break savegame once
16:09:49  <andythenorth> and there is potentially more savegame breaking to do
16:09:51  <planetmaker> and if you moved the savegame break to 0.8
16:09:56  <andythenorth> that would be fine
16:10:01  <andythenorth> we'd just need to drop some revisions
16:10:14  <andythenorth> I'd be happier porting 0.6.4 and bringing the changes since then across later
16:10:27  * andythenorth has to go out for an hour or so...
16:10:51  <andythenorth> hmm
16:10:51  <planetmaker> andythenorth, so... forgetting all changes since 0.6.4 and converting that?
16:10:57  <andythenorth> there are about 100 revisions since 0.6.4
16:11:06  <Terkhen> that sounds like a lot of work :/
16:11:16  <planetmaker> graphics will be kept
16:11:27  *** frosch123 has joined #openttdcoop.devzone
16:11:31  <andythenorth> I'd have to look in more detail
16:11:42  <Terkhen> ok :)
16:11:43  <andythenorth> the attraction is that we then get a better source of bug reports for the change
16:11:58  <planetmaker> of course
16:12:00  <andythenorth> otherwise, FIRS does not have that big a user base
16:12:14  <andythenorth> and adoption of new versions is slow for obvious reasons
16:12:25  <andythenorth> anyway, bbl
16:13:03  <Terkhen> enjoy
16:13:21  <planetmaker> hm, ... lots of changes.... most is code
16:13:26  <Yexo> a lot of the manual work of the conversion will be to rename string names so they make sense again
16:13:35  <planetmaker> yes
16:13:39  <Yexo> we could do a 0.6.5 with minimal manual work too see if that succeeds
16:13:56  <Yexo> than completely redo the conversion in 0.7
16:14:11  <planetmaker> well... possible.
16:14:16  <Terkhen> that makes sense
16:14:19  <Yexo> most of it is automatic and the few bits that are not will show up in the hg log, so they can be repeated easily
16:14:44  <planetmaker> alright :-)
16:15:09  <planetmaker> let's branch the 0.6.4 then
16:16:03  <Brot6> OpenGFX+ Industries - Code Review #2092 (Closed): low max value (Terkhen) @ http://dev.openttdcoop.org/issues/2092#change-6682
16:21:55  <Terkhen> planetmaker: I would like to release a new version of opengfx+ industries soon
16:22:32  <Terkhen> right now I'm on a break from my project, but I don't know how long will it last, and once it finishes I will not be doing much until probably the end of june
16:24:45  <planetmaker> Yes, so would do I. It could ship as of now. But one could change a few things, though, before: the ground awareness of the water well. The rest is - afaik - not critical in any way
16:25:05  <planetmaker> maybe the rubber plantation and fruit plantation could get snowy ground on snow... but...
16:26:24  <Terkhen> the only remaining thing that might be important to have is #2612
16:26:24  <Brot6> Terkhen: #2612 is http://dev.openttdcoop.org/issues/show/2612 "OpenGFX+ Industries - Bug #2612: Some industries and cargos use the same colours. - #openttdcoop Development Zone"
16:26:41  <Terkhen> then again, I had played for an hour already when I noticed
16:28:15  <planetmaker> well, it's a bug, but anything but crucial
16:29:36  <Terkhen> yes
16:29:56  <Terkhen> should we release once I commit conflicting industries then?
16:30:56  <planetmaker> I'd not mind. And we need to keep some bugs for the players to find ;-)
16:31:13  <planetmaker> If you want to release today, you'll have to do that, I'll be off for sports soon
16:31:45  <Terkhen> okay :)
16:32:00  <Terkhen> can you please point me to your script again?
16:32:49  <planetmaker> uh, the make_changelog one?
16:33:16  <planetmaker> http://devs.openttd.org/~planetmaker/patches/make_changelog.sh
16:33:34  <planetmaker> uff... I thought I pasted it last time. Then it would have been difficult, I only have it at home
16:37:45  <Terkhen> yes, thank you :)
16:54:16  <andythenorth> so if we branch at 0.6.4 equivalent revision...
16:54:26  <andythenorth> we can port changes from trunk?
16:54:32  <andythenorth> then merge the branches back later?
16:54:47  <andythenorth> the nfo will get binned
16:54:58  <andythenorth> but the psds / pcx are portable
16:55:12  <planetmaker> they'll be untouched
16:55:29  <Yexo> andythenorth: my last idea was: do an automatic conversion of 0.6.4 and only manually fix any bugs / omissions
16:55:56  <Yexo> when that release is out for a while (so it's tested), convert the trunk version at that time and fix bugs / omissions again
16:56:10  <Yexo> only for that second conversion make new templates for the nml code
17:03:10  <andythenorth> so if (a) is the automatic conversion and (b) is the templated version
17:03:25  <andythenorth> we would be able to compare output of (a) and (b)?
17:03:35  <Yexo> yes
17:03:52  <planetmaker> well, (b) would also be automatic. But MUCH more hand-tuned afterwards
17:04:27  <Yexo> no, (b) is manually created from (a)
17:04:31  <planetmaker> if you looked ever at opendune: (a) would be like right after de-compilation and (b) like where it is now
17:05:06  <Yexo> well, yes :)
17:06:07  <andythenorth> http://www.joelonsoftware.com/articles/fog0000000069.html
17:06:08  <Webster> Title: Things You Should Never Do, Part I - Joel on Software (at www.joelonsoftware.com)
17:06:08  <planetmaker> Yexo, I'd do an automatic conversion for trunk, too. Or you want to write-in all nfo changes since 0.6.4?
17:06:10  <andythenorth> http://www.joelonsoftware.com/articles/fog0000000348.html
17:06:11  <Webster> Title: Rub a dub dub - Joel on Software (at www.joelonsoftware.com)
17:06:57  <Yexo> planetmaker: yes, so we have (a1) =automatic conversion of 0.6.4, (a2) = automatic conversion of trunk and (b) = manual adjustment with templates for (a2)
17:07:25  <planetmaker> yes. Glad we agree :-)
17:07:37  <Yexo> andythenorth: that's a very good article, but not relevant here
17:07:45  <Yexo> a conversion between languages is hardly a "complete rewrite"
17:07:58  <Yexo> we're not starting to code from scratch
17:08:57  <Terkhen> it's more like a translation from chinese to russian :P
17:09:55  <andythenorth> the second article is more relevant I think
17:10:06  <andythenorth> iirc he upgraded code but didn't make any functional changes
17:10:15  <andythenorth> 1:1 retention of features
17:10:25  <andythenorth> assuming he did it correctly :P
17:10:37  <Terkhen> venga
17:10:48  <Terkhen> ups, sorry :)
17:11:15  <andythenorth> anyway, my main concern is making best use of current installed base, such as it is
17:11:32  <andythenorth> asking people to apply this to existing savegames should be good
17:11:48  * Terkhen agrees
17:13:19  <andythenorth> although I'm not sure how to inform users what the change is, nor how to report bugs
17:13:29  <andythenorth> bananas is not of use for either
17:13:42  <Yexo> same as always, report bugs in the forum topic
17:13:51  <Terkhen> big refactoring of code that might but should not result in changes of behaviour
17:14:04  <Terkhen> why isn't bananas of use?
17:14:13  <andythenorth> how do we tell users what's changed?
17:14:20  <Yexo> how do you tell that normally?
17:14:28  <Yexo> ie what is the difference with a 'normal' release?
17:14:38  <andythenorth> I don't tell that normally for bananas users
17:14:39  <Terkhen> normal releases have visible changes
17:14:46  <andythenorth> there isn't any way to tell bananas users
17:14:50  <Yexo> so why would there be a reason to tell them now?
17:15:58  <andythenorth> there probably isn't it
17:16:09  <andythenorth> but it reduces the validity of my argument about installed base ;)
17:16:20  <andythenorth> maybe we just do a bundle server + forums release
17:16:41  <andythenorth> FIRS already has a mixed reputation, I don't want to make that worse by inflicting a possibly significantly broken 0.6.5 on bananas users
17:16:46  <Yexo> let's not get too far ahead and first make it work, ok?
17:16:56  <andythenorth> yarp
17:17:00  <andythenorth> thought you might say that ;)
17:17:08  <andythenorth> anyway, I think I get ti
17:17:10  <andythenorth> it /s
17:17:26  <andythenorth> by doing 0.6.4 -> 0.6.5, we test not FIRS, but the conversion script?
17:17:37  <Yexo> I agree with you that we have to be careful to make sure 0.6.5 is not significantly broken
17:17:46  <Yexo> basically yes
17:18:00  <planetmaker> but getting it on bananas IMHO should be part of it
17:18:00  <andythenorth> then if the script proves good, we can reuse it on trunk
17:18:07  <Yexo> yes, exactly
17:18:08  <Brot6> fish: update from r622 to r624 done - http://bundles.openttdcoop.org/fish/nightlies/r624
17:18:32  <Brot6> ogfx-industries: update from r71 to r74 done - http://bundles.openttdcoop.org/ogfx-industries/nightlies/r74
17:18:40  <andythenorth> planetmaker: I think bananas is just a question of timing
17:18:47  * andythenorth advocates patience (unusually)
17:18:48  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r750), 32bpp-extra (r40), ai-admiralai (r75), ai-aroai (r39), ailib-common (r21), ailib-direction (r17), ailib-list (r32), ailib-string (r29), ailib-tile (r16), airportsplus (r73), basecosts (r25), belarusiantowns (r8), bros (r52), chips (r138), comic-houses (r71), firs (r1989), frenchtowns (r6), german-townnames (r33), grfcodec (r828), grfpack (r279), heqs (r605),
17:18:48  <Brot6> indonesiantowns (r41), manindu (r7), metrotrackset (r56), narvs (r37), newgrf_makefile (r285), nml (r1334), nutracks (r186), ogfx-landscape (r67), ogfx-rv (r80), ogfx-trains (r239), ogfx-trees (r42), opengfx (r661), openmsx (r97), opensfx (r97), smts (r19), snowlinemod (r49), spanishtowns (r10), sub-opengfx (ERROR r666), swedishrails (r202), swisstowns (r22), transrapidtrackset (r15), ttdviewer (r34), ttrs (r36), worldairlinersset (r671)
17:18:53  <andythenorth> release to bundles server first
17:18:59  <andythenorth> test for a few weeks
17:19:03  <andythenorth> then release to bananas
17:19:24  * andythenorth has to go out again :)
17:19:25  <planetmaker> uhm... that's a bit pointless. There's no need to release then to the bundles server either
17:19:35  <andythenorth> planetmaker: ach, we'll figure it out in a bit
17:19:49  <Brot6> sub-opengfx: compile of r666 still failed (#2586) - http://bundles.openttdcoop.org/sub-opengfx/nightlies/ERROR/r666
17:19:52  <andythenorth> there's no point releasing a potentially significantly buggy major change to bananas ;)
17:20:05  <andythenorth> it will just further damage the reputation of FIRS
17:20:07  <planetmaker> the point is: test before any release
17:20:22  <planetmaker> as carefully as usually (should be done)
17:20:26  <Yexo> we won't put 0.6.5 out before we've done proper testing ourself
17:20:37  <planetmaker> ^
17:20:45  <andythenorth> well - as you said - that can wait for now ;)
17:20:55  <andythenorth> the key thing is to use 0.6.4 for the test?
17:20:58  * andythenorth bbl
17:21:15  *** andythenorth has quit IRC
17:41:38  * planetmaker is off for sports. See you much later or tomorrow
18:00:42  <Brot6> OpenGFX+ Industries - Revision 75:8a997ecdda74: Add: Conflicting industries for the Factory, the ... (Terkhen) @ http://dev.openttdcoop.org/projects/ogfx-industries/repository/revisions/8a997ecdda74
18:29:11  <Terkhen> hmm... translations
18:48:56  <Brot6> OpenGFX+ Industries - Revision 76:85d2fa21a547: Change: Update Spanish translation. (Terkhen) @ http://dev.openttdcoop.org/projects/ogfx-industries/repository/revisions/85d2fa21a547
18:49:52  <Brot6> OpenGFX+ Industries - Revision 77:f1a3829e6ae8: Doc: Update changelog. (Terkhen) @ http://dev.openttdcoop.org/projects/ogfx-industries/repository/revisions/f1a3829e6ae8
18:50:29  <Terkhen> planetmaker: ^ only your translation is missing; once it is done we can release
18:50:33  <Terkhen> and as always, there is no need to hurry :P
18:58:50  *** LordAro has joined #openttdcoop.devzone
19:10:41  *** andythenorth has joined #openttdcoop.devzone
19:11:05  <andythenorth> hmm
19:11:16  <andythenorth> has #openttd been blocked for some reasons?
19:11:20  <andythenorth> I can't get in
19:11:30  <andythenorth> I can get in here on oftc.net
19:13:00  <LordAro> need to register/login
19:13:20  <andythenorth>  I (re)registed already a few weeks ago
19:13:30  <andythenorth> but thanks ;)
19:14:07  <andythenorth> I'll just not bother
19:22:17  <Terkhen> andythenorth: there were a lot of spambots joining and leaving, in those moments the channel is set to "registered only"
19:22:38  <andythenorth> ok
19:24:38  <andythenorth> any more FIRS-> nml chat?
19:29:14  <Terkhen> nope
19:37:43  *** LordAro has quit IRC
20:37:22  <planetmaker> good evening. Hm.. translation missing. Again I shall see to that now. Forgot about that, thanks for reminding, Terkhen
20:37:51  <Terkhen> ok :)
20:38:44  <Yexo> planetmaker: if #2614 is also present in 0.6.4 that'll lead to problems with the conversion
20:38:44  <Brot6> Yexo: planetmaker: #2614 is http://dev.openttdcoop.org/issues/show/2614 "FIRS Industry Replacement Set - Bug #2614: Translation framework broken when not all texts are translated - #openttdcoop Development Zone"
20:39:17  <Yexo> could you take a look at that?
20:58:24  <Brot6> FISH - Revision 625:660a4f8b7c55: Change: increase capacity of small utility tug (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/660a4f8b7c55
21:02:34  <planetmaker> I will, Yexo
21:04:40  *** andythenorth has quit IRC
21:06:29  <Yexo> thanks
21:06:58  <Yexo> I could change some code in firs to make it easier to convert, would that be ok or do you think it'll lead to problems later on?
21:07:35  *** frosch123 has quit IRC
21:08:28  <planetmaker> talking to me and concerning the translations?
21:08:39  <Yexo> yes and no
21:09:28  <planetmaker> :-) I would not mind that
21:09:40  <planetmaker> What would you change? Why and in what way?
21:10:04  <Yexo> make some of the action7/9 'properly' nest
21:10:17  <planetmaker> Ultimately I'd consider andy the authority to decide on that. I doubt he'd mind either
21:10:49  <Yexo> currently there is code like this: suppose A, B and C are consecutive sprite numbers. A checks condition and skips B if there is an error
21:11:05  <Yexo> B unconditionally skips C, C is the actually actionB (=error) action
21:11:25  <Yexo> this is hard to convert automatically
21:11:29  <planetmaker> ah, that code.
21:11:40  <planetmaker> I wrote it. At least partially ;-)
21:11:49  <planetmaker> it's short and easy to understand in nfo
21:12:12  <planetmaker> But I don't mind other solutions for all those grf and compatibility checks
21:13:20  <Yexo> I think some of the checks are wrong, like //Terminate if TTDP 24 * 6	 09 9A 01  00 00
21:13:44  <Yexo> hmm, nvm
21:13:48  <Yexo> that is a wrong conversion
21:27:50  <planetmaker> hm. fantastic. saw mill and lumber mill have the same translation in German. And the current translation of lumber mill in trunk is basically bogus :-(
21:32:59  <Brot6> OpenGFX+ Industries - Revision 78:1abba00e548f: Change: Update German translation (planetmaker) @ http://dev.openttdcoop.org/projects/ogfx-industries/repository/revisions/1abba00e548f
21:49:20  <Terkhen> planetmaker: the same happens with spanish, I decided to ignore the issue for now :P
21:49:35  *** ODM has quit IRC
21:49:44  <Terkhen> Yexo, planetmaker: IIRC FIRS string defines are prepared in a way that, if one of them is not defined for a translation, it will use the english string instead
21:49:54  <Terkhen> I would need to check the code to be sure, though
21:52:44  <Yexo> Terkhen: that might have been the intention, but that's not the case
21:52:51  <Yexo> check sprite 507 (in firs 0.6.4)
21:53:16  <Terkhen> I can't check right now but... bad :(
21:53:32  <planetmaker> Hm... not sure. I think it does not define the string then (which actually is the better thing). But... I'm too tired to solve that now. I'll have to defer that tomorrow, I fear. But what I saw earlier today, it looks like a bug which needs fixing
21:54:41  <Terkhen> but given that the translation framework will not be necessary with nml, the fix could be just making sure that all translated strings appear correctly in the compiled grf so the script can pick them up
21:55:28  <Yexo> added a fix for the single case I found to the issue tracker
21:55:39  <Yexo> but the lang file needs checking if there are other cases
21:55:52  <Brot6> FIRS Industry Replacement Set - Bug #2614: Translation framework broken when not all texts are tr... (yexo) @ http://dev.openttdcoop.org/issues/2614#change-6683
21:56:00  <planetmaker> there might. I guess you moved the action7 into the #ifdef?
21:57:21  <planetmaker> hm, yes. I guess EVERY string will need that change. :S
21:58:37  <Yexo> most strings don't have an action7
21:59:18  <Yexo> it might actually be the only case, since for TEXT_INFO_5T_PER_8T_SUGARCANE and TEXT_INFO_5T_PER_8T_SUGARBEET it's already done
22:00:26  <Yexo> good night
22:00:37  <Terkhen> good night from here too :)
22:00:39  <planetmaker> good night :-)
23:26:06  *** Lakie has quit IRC

Powered by YARRSTE version: svn-trunk